5 Managing Events, Alerts, and Incidents
5.1 Events
An event is a significant occurrence that indicates a potential problem. When a metric threshold value is reached, a metric alert is raised. A metric alert is a type of event. An alert can also be generated for various target availability states.
Event Types
Typically, key event types used in Enterprise Monitoring are:- Metric Alert: A metric alert event is generated when an alert occurs for a metric on a specific target or metric on a target and object combination, such as Lag Exceeding a Specified Threshold Value.
- Target Availability: The Target Availability Event represents a target's availability status. For example: Up, Down, Agent Unreachable, or Blackout. For more information on all the targets available in Oracle GoldenGate, see Supported Target Types.
5.2 Metric Data and Alerts
Metric data refers to the collection of data that changes frequently. You can create alerts on the metric data. Oracle GoldenGate delivers predefined metric types and default collection times for each target type.
- Checkpoint Position
- Name
- Status
- Start Time
- End of File
- Lag (Sec)
- Total Inserts
- Delta Inserts
- Total Deletes
- Delta Deletes
- Total Truncates
- Delta Truncates
- Total Operations
- Delta Operations
- Delta Operation Per seconds
- Total Executed DDLs
- Delta Executed DDLs
- Total Discards
- Delta Discards
- Total Ignores
- Delta Ignores
- Last OGG Checkpoint Timestamp
- Last Processed Timestamp
- Delta Row Fetch Attemps
- Delta Row Fetch Failures
- Total Row Fetch Failures
For more information on the metric data, see Extract and Replicat. The metric data collected is saved to the Management Repository and is compared to the predefined thresholds for each target. If a threshold is reached, then the system generates an alert. The Incidents are displayed on each of the target's homepage.
5.3 Incidents and Alerts
An incident is a unit containing a single, or closely correlated set of events that identify an issue that needs administrator attention. Although incidents can correspond to a single event, incidents more commonly correspond to groups of related events.
The Oracle Enterprise Manager provides various options to respond to Incidents. Administrators can be notified automatically when an alert triggers and can set up corrective actions to resolve an alert condition automatically.
You can set metric alerts and also generate alerts for various target availability states. This topic details the following:5.3.1 Setting Metric Alerts and Incidents for Extract and Replicat
If you want to set alerts of metric status values, the following are the status values for Extract and Replicat.
- Registered - 2
- Starting - 3
- Running - 7
- Stopping - 8
- Stopping Forcefully - 9
- Stopped - 10
- Stopped Forcefully - 11
- Abended - 12
- Killed - 13
- Unresponsive - 16
Oracle recommends to set alerts for target availability to monitor the status of the targets.
5.3.2 Setting Incidents and Alerts for Oracle GoldenGate Target Availability
You need to set alerts on Target Availability of Oracle GoldenGate targets to get notified when there are any issues with these targets.
- On the Home page, click Setup, select Incidents, and then click Incident Rule to display the Incident Rules - All Enterprise Rules page.
- Click Create Rule Set....
- Enter a Name, for example Incident management rule set for Target Availability and click Save.
- In the Target area, select All Targets of types, and select the target type from the adjacent drop-down.
- In the Rules area, click Create... to display the Select Type of Rule to Create dialog box.
- Select Incoming events and updates to events and click Continue to display the Create New Rule: Select Events page.
- Select Target Availability from the Type drop-down list and click Next to display the Create New Rule: Add Actions page.
- Click Add to display the Add Conditional Actions page, select Always execute the actions.
- Under Send Notifications, expand Basic Notifications, and enter email IDs in E-mail To and E-mail Cc to assign recepients for notifications. These email IDs can belong to the users of the Enterprise Manager.
- Click Continue to view the Action Summary in the Create New Rule: Select Events page.
- Click Next to display the Create New Rule: Specify Name and Description page, where a new Rule, for example, rule 166 is displayed. You can either specify a rule name or click Next to accept the pre-specified name to display the Create New Rule: Review page.
- Click Continue and then click Save to save the new
rule.
In this example, a rule 166 has been successfully created and added to the current rule set. Incident management rule set for Target Availability is the incident rule set that has been set on Target Availability of the selected targets, which will trigger alert and send emails to the recepients specified in case of issues or events with these targets.
5.4 Alerts on Home Page
The Oracle GoldenGate Home page displays all the incidents that are generated. An alert is generated when a metric threshhold is reached. The most recent alerts are listed first.
- On the OGG Home page, click the number (Critical or Warning) under Incidents to display the Incident Manager.
- Click an alert message to view all the details about the selected metric in the alert.